This prospective longitudinal study is designed to identify: 1) late out- comes and consequences related to HIV disease progression, treatment and prophylactic intervention, and/or interaction of HIV disease and therapy, and 2) late treatment effects among perinatally treated infants proven to be uninfected. Specific hypotheses being addressed include the following: a) Initial combination therapy as compared to monotherapy (e.g. following a randomization treatment protocol such as ACTG 152) will result in significantly increased long-term survival and improved physical growth and quality of life; b) Initial use of zidovudine in infancy (0-24 months) either in combination or as monotherapy (e.g. ACTG 152) will result in significantly higher mean neurocognitive scores at age 4 and 7 years compared to initial use of other antiretroviral therapies that do not include zidovudine; c) There will be different rates of specific organ toxicity at ages 4, 7, 15 and 20 years among children and adolescents by treatment regimen; d) There will be no significantly increased prevalence of birth defects, neoplasma, or late organ toxicity among ultimately uninfected infants exposed to ZDV in utero compared to similar pediatric populations in the USA who were not exposed in utero to ZDV; and e) Advances in anti-HIV therapy and improved supportive care over time will result in significantly increased survival, improved growth and development, decreased morbidity and improved quality of life for successive cohorts of children involved in ACTG treatment protocols.